What affects electron mobility in semiconductors and metals?

[Chiz]

Hi there,

i've got a question about the mobility of the electrons in semiconductors and metalls.
in a semiconductor the mobilities are ~ 1000 cm²/Vs. In metals the mobilities are in the range of 50 cm²/Vs. what's the origin of that great difference?

 

[johng23]

Physically, carriers will accelerate in an applied field until they scatter off of some obstacle, at which point they start to accelerate again and end up having some average velocity (drift velocity) when acted on by the field. The mobility relates this velocity to the magnitude of the field, so its essentially a measure of how often carriers scatter. I think the reason that semiconductors have such high mobility is because the intrinsic carrier concentration is so low that electrons don't scatter off of each other. As the figure below shows, at high doping levels the carrier density approaches that of a metal, and the mobility drops down towards that 50 cm²/Vs value that you mentioned.

 

[Chiz]

But in metals the electron-electron interaction is very small because of the pauli principle. that's explained in the book of Ibach and Lüth "Solid state physics". there's written that you can neglect that effect and can treat the electrons as non-interacting particles. so in my opinion the electron electron interaction can't be the reason for that great difference in the mobilities.

 

[johng23]

hmm ok it was just a guess which seemed slightly plausible. Is there any difference in the phonon population of semiconductors and metals?
Let me know if you come up with anything, I'm also interested in this question.